[Mlir-commits] [mlir] [mlir][arith] Add LUT-based lowering of arith.extf for FP8 to F32. (PR #196321)

Krzysztof Drewniak llvmlistbot at llvm.org
Thu May 7 09:07:53 PDT 2026


krzysz00 wrote:

1. This should live in arith/Transforms
2. Since we're here, I'll note that we could also implement truncation using the same lookup table and a binary search - and probably should.

https://github.com/llvm/llvm-project/pull/196321


More information about the Mlir-commits mailing list